List your company

Multi-hop Reasoning on Housing QA

82.67AccuracyTOTAL
56.6763.4270.1776.92Oct 8, 2025

Evaluation results

MethodLinks
TOTALModel=Claude, Prompting Method=TOTAL2025.1082.67
CIC + COTModel=Claude, Prompting Method=CIC + COT2025.1075
TOTALModel=Gemini, Prompting Method=TOTAL2025.1074.33
CICModel=Claude, Prompting Method=CIC2025.1071.67
CIC + COTModel=Gemini, Prompting Method=CIC + COT2025.1070.33
TOTALModel=GPT, Prompting Method=TOTAL2025.1070
CICModel=Gemini, Prompting Method=CIC2025.1068.33
CIC + COTModel=GPT, Prompting Method=CIC + COT2025.1066
CICModel=GPT, Prompting Method=CIC2025.1064.33
COTModel=GPT, Prompting Method=COT2025.1061.33
NAÏVEModel=Claude, Prompting Method=NAÏVE2025.1060.33
NAÏVEModel=GPT, Prompting Method=NAÏVE2025.1060.33
COTModel=Gemini, Prompting Method=COT2025.1058.67
NAÏVEModel=Gemini, Prompting Method=NAÏVE2025.1058
COTModel=Claude, Prompting Method=COT2025.1057.67